Announcing Trading Strategy protocol

In this blog post, we announce Trading Strategy, a decentralised protocol for algorithmic trading.

Vision

After working on and analysing blockchain, crypto, and decentralised finance for many years, we feel that the time has come for fully programmed quantitative hedge funds.  As decentralised finance (DeFi) matures, more and more financial services can be constructed purely in code, whereas before they existed as traditional and bureaucratic paper-pusher institutions.

Today, decentralised markets exist, have matured enough, and have various infrastructures supporting them, from lending services to governance. However, the use of decentralised exchanges and other DeFi services has not been an intuitive or easy experience for even many experienced cryptocurrency traders and investors. This is due to the very different operational and mental models that these services have implemented compared to the more traditional centralised services. Though young, there already exists considerable volumes, different assets, and various investment opportunities that can be accessed in a purely programmed manner, directly on-chain - no paper agreement or negotiation talks are needed to enter into a trading position.

What is algorithmic trading?

Algorithmic trading is a derivative of technical analysis; taking trading positions based on pure mathematics and data. Algorithmic trading is part of quantitative finance, the opposite of value investing where trading decisions are made based on fundamentals. Algorithmic trading provides a systematic approach to trading compared to methods based on trader intuition or instinct. Whereas technical analysis often aids humans to take trading positions, in its purest form in algorithmic trading a trading program follows a set of trading rules and independently executes trades on the market 24/7.

Algorithmic trading can follow several different types of strategies, for example, directional strategies anticipating market moves based on trends or mean reversion or market neutral strategies where the algorithm seeks to make a risk-free profit over arbitration and dislocations over different markets.

Towards decentralised quantitative funds

Algorithmic trading already exists in centralised markets (stock), centralised cryptocurrency exchanges (Binance, Coinbase, Kraken, others), and also on decentralised exchanges like Uniswap. However, these activities are carried out by actors specialised in quantitative finance, hedge funds, and proprietary traders. Although the markets themselves are becoming decentralised, the investment opportunity into decentralised trading itself has not been decentralised yet.

The first steps in this direction have already been taken by Enzyme Protocol and dHedge.  They look to decentralise the fund infrastructure and the back office work: the act of investing into a fund, accounting fees and redemption.

TradingStrategy.ai takes a step further and decentralises the role of an investment manager - the person who is going to decide what trades to take and execute them.

Total addressable market

The advantage of using software to automate trading is profound, mostly in its effect on lowering operational costs but also by removing human emotions from the actual trading execution. The software does not have feelings, it can compare and analyse loads of different market indicators in a split second, and does not make mistakes under stress or euphoria like human traders. Unsurprisingly, quantitative funds commence the highest trading volume on the US stock markets of all different trader types.

To gauge the potential of the “decentralised investment manager”, the worldwide open-ended funds had $63 trillion under management by the end of 2020. Assuming the share of quantitative trading increases to 50% and everything will be tokenised over the next decade or so, the total addressable market would be around 30$ trillion, in today’s dollars, counted as assets under management (AUM).

Unique benefits of the decentralised protocol

Smart contracts and decentralised finance enable so-called non-custodial business models. The investor never loses control of their assets (money, trading positions) and can redeem assets any second, no questions asked. In the Trading Strategy protocol the investment manager, which is a piece of program code, cannot misappropriate the client’s assets. Overcharging, disagreements regarding fees or late payments simply cannot exist because those are governed by smart contracts, not fund managers or office workers.

Decentralised exchanges, where the trading happens, do not have counterparty risk. A decentralised exchange cannot go bankrupt. Because there is no credit risk and all settlements happen in real-time, assessing the systematic risk is much easier, the overall risk is lower.

As the first of its kind, the Trading Strategy protocol runs its own network of decentralised oracles that are specialised in algorithmic trading and can make trading decisions based on gigabytes of market and price data. The protocol is unique from its market and price data collection point of view as we are collecting, cleaning, and processing the raw data directly from the blockchains. This removes dependencies to the third party APIs and mitigates data supply risks.

Protocol stakeholders

The Trading Strategy protocol stakeholders will be investors, strategy developers, oracle network node operators and the protocol development team. We believe there is an opportunity to combine important value-adds from all of the groups around the protocol.  Each of the parties bring their own assets to the ecosystem whether they are monetary, knowledge-based, or technical capabilities.

Furthermore, we see plenty of integration opportunities with decentralised exchanges, layer one and two blockchains, wallets, aggregators and other liquidity providers.  

Business model

The business models of existing quantitative funds can be translated directly to decentralised trading strategies. The fee models can vary between 40% performance fees (top solo traders) to 2% management fee / 20% performance fee, a standard hedge fund structure.

This business model has already been successfully proven in the decentralised finance space, as the largest yield farming protocol, Yearn Finance is currently taking 2/20 fees

The proceeds are distributed among four different stakeholders: the algorithm developer, the protocol treasury (“development fund”), marketing referrers and decentralised oracle network operators. For more examples of fee structures, see the documentation.

Below you find the protocol revenues for popular DeFi protocols from Token Terminal.

Comparison

Here is a brief comparison and key differentiators between the Trading Strategy protocol and its closest benchmarks.

Quantitative funds operate in and obey the rules of traditional finance. Their correctness is enforced by law. Activities are governed by regulation, audits, and litigation on disagreements. Quantitative funds can employ trading strategies everywhere from simple directional candlestick counting to complex machine learning-based statistical algorithms.

Robo-advisors, “trading bots” and trading signal services offer semi-automated or automated trading for centralised exchanges. They charge a flat monthly fee regardless of their success. There is no transparency and these services may easily screw their investors: front-running, adding additional margins/spread, selling order flow, or leaking API keys.

There already exist on-chain trading strategies: yield farming, option strategies, and similar passive strategies. The computational limit of these strategies is that they execute all logic on-chain. A smart contract transaction can analyse a few kilobytes at most.  Thus, these strategies often centralise around compounded liquidity mining where another protocol is rewarding liquidity providers with airdropped equity tokens.

Trading Strategy protocol operates its own decentralised oracle network that analyses trading signals, executes strategies, and makes rebalancing decisions off-chain. Only the consensus on the next trading move and asset management happen on-chain. This combines high-performance off-chain computing for the trading decisions to the safe and transparent on-chain asset management and trade execution. The decentralised oracles make trading decisions based on gigabytes of market and price data.

Product-market fit

Can decentralised trading algorithms yield profit to investors? This question will determine the success of the Trading Strategy protocol. The answer is very likely.

Trading Strategy will support both public (open source) and private (closed source) trading strategies.

• Public strategies are school book strategies. While these strategies are well-known, most of the traders do not employ them, especially on under-developed decentralised trading markets. These strategies can typically yield anything between 5% - 25% yearly depending on the market conditions.
• Private strategies should be able to yield performance comparable to quantitative hedge funds, as they are likely developed by the same actors with the same algorithms. Private strategies are partially verified or solely rely on the decisions of the operator. Private directional strategies are known to be able to outperform Bitcoin/Ethereum buy-and-hold strategies with considerably less drawdown.

Another way to look at the success problem: there exist decentralised lending markets worth $30B (September, 2021). Somebody is borrowing all this money which means borrowers’ business activities must be profitable. Most if not all of this borrowed money goes into cryptocurrency trading. Thus, these traders must outperform the borrowing rate, which in the case of decentralised finance is the variable interest rate on US dollar you can get in Aave, Alpha Homora, and Compound lending pools (around 1% - 7%). Assuming a school book algorithm reaches at least partially similar results, we should outperform this risk-free rate with the introduction of a well-managed drawdown.

Below is a risk-reward matrix of different Defi trading options

As long as the Trading Strategy protocol based trading algorithms are more rewarding than risk-free lending rate and less risky than buy-and-hold, the protocol should be able to offer an attractive product for DeFi investors.

Below is another crypto-related investment benchmark, based on the rough market understanding. Here we include the deployable funds in the comparison: selective funds like early crypto VCs and HFT traders will hit their limit of running out of investable opportunities fast, whereas more generic trading strategies can absorb, if not unlimited, at least a very high amount of capital. The largest Uniswap v3 markets are measured in billions - you can execute million-dollar trades without practical slippage. With the advent of DyDx and other decentralized futures markets, we see decentralised derivative markets being even more scalable.

Next steps

We will start deploying the first strategies on EVM chains soon, available as a private beta. This is uncharted waters: we want to limit any potential realisation of technical risk for investors who understand the implications of “software still in development”.

We are currently looking for partners in the decentralised finance space who either want to

1. Help us develop trading algorithms,
2. Integrate our strategies to their exchanges, and
3. Offer Trading Strategy protocol directly to their users, through a wallet or other similar service in exchange for the referral fees.

Please shoot us a business collaboration enquiry if you belong to any of these categories.

In the meanwhile, everyone else is free to join our Discord and explore our draft documentation.

The post feature image by Maxim Hogman.